Collision-Free Trajectory Planning and control of Robotic Manipulator using Energy-Based Artificial Potential Field (E-APF)

TL;DR

This paper introduces an Energy-based Artificial Potential Field (E-APF) framework for robotic trajectory planning that combines position and velocity potentials to avoid local minima and oscillations, ensuring smooth, collision-free, and time-efficient motion.

Contribution

The novel E-APF framework integrates velocity-dependent potentials with traditional methods, improving dynamic adaptability and obstacle avoidance in robotic trajectory planning.

Findings

Successfully avoids local minima and oscillations in simulations

Produces smooth, collision-free trajectories in cluttered environments

Enhances time efficiency and motion quality of robotic manipulators

Abstract

Robotic trajectory planning in dynamic and cluttered environments remains a critical challenge, particularly when striving for both time efficiency and motion smoothness under actuation constraints. Traditional path planner, such as Artificial Potential Field (APF), offer computational efficiency but suffer from local minima issue due to position-based potential field functions and oscillatory motion near the obstacles due to Newtonian mechanics. To address this limitation, an Energy-based Artificial Potential Field (APF) framework is proposed in this paper that integrates position and velocity-dependent potential functions. E-APF ensures dynamic adaptability and mitigates local minima, enabling uninterrupted progression toward the goal. The proposed framework integrates E-APF with a hybrid trajectory optimizer that jointly minimizes jerk and execution time under velocity and acceleration constraints, ensuring geometric smoothness and time efficiency. The entire framework is validated in simulation using the 7-degree-of-freedom Kinova Gen3 robotic manipulator. The results demonstrate collision-free, smooth, time-efficient, and oscillation-free trajectory in the presence of obstacles, highlighting the efficacy of the combined trajectory optimization and real-time obstacle avoidance approach. This work lays the foundation for future integration with reactive control strategies and physical hardware deployment in real-world manipulation tasks.